qanom-seq2seq-model-joint

A Seq2Seq model for QANom parsing

This is a t5-small pretrained model, fine-tuned jointly on the tasks of generating QASRL and QANom QAs.

"QANom" stands for "QASRL for Nominalizations", which is an adaptation of QASRL (Question-Answer driven Semantic Role Labeling) for the nominal predicates domain. See the QANom paper for details about the task. The QANom Dataset official site is a Google drive, but we also wrapped it into a Huggingface Dataset, which is easier to plug-and-play with (check out our HF profile for other related datasets, such as QASRL, QAMR, QADiscourse, and QA-Align).

Demo

Visit our demo for interactively exploring our model!

Usage

The model and tokenizer can be downloaded as simply as running:

import transformers
model = transformers.AutoModelForSeq2SeqLM.from_pretrained("kleinay/qanom-seq2seq-model-baseline")
tokenizer = transformers.AutoTokenizer.from_pretrained("kleinay/qanom-seq2seq-model-baseline")

However, the model fine-tuning procedure involves input preprocessing (marking the predicate in the sentence, T5's "task prefix", incorporating the predicate type and/or the verbal form of the nominalization) and output postprocessing (parsing the sequence into a list of QASRL-formatted QAs).
In order to use the model for QANom parsing easily, we suggest downloading the pipeline.py file from this repository, and then use the QASRL_Pipeline class:

from pipeline import QASRL_Pipeline
pipe = QASRL_Pipeline("kleinay/qanom-seq2seq-model-joint")
pipe("The student was interested in Luke 's <predicate> research about sea animals .", verb_form="research", predicate_type="nominal")

Which will output:

[{'generated_text': 'who _ _ researched something _ _ ?<extra_id_7> Luke', 
  'QAs': [{'question': 'who researched something ?', 'answers': ['Luke']}]}]

You can learn more about using transformers.pipelines in the official docs.

Notice that you need to specify which word in the sentence is the predicate, about which the question will interrogate. By default, you should precede the predicate with the <predicate> symbol, but you can also specify your own predicate marker:

pipe("The student was interested in Luke 's <PRED> research about sea animals .", verb_form="research", predicate_type="nominal", predicate_marker="<PRED>")

In addition, you can specify additional kwargs for controling the model's decoding algorithm:

pipe("The student was interested in Luke 's <predicate> research about sea animals .", verb_form="research", predicate_type="nominal", num_beams=3)